Shiny Things

The "Shiny Things" thread on HWZ Money Mind is a goldmine of valuable investment information. I have benefited tremendously and I highly encourage any investor, new or experienced, to slowly digest the wealth of goodies within that thread.

While I have always known the benefits of low cost ETFs, the discussions further strengthened my perception on them, and more importantly on asset allocation and re-balancing.

The best thing is you get concrete details on the exact ETF to purchase (to cut taxes to minimum), and numerous real-life examples on portfolio allocation.

Below are some copy and paste bits of information that I feel will be useful for my future reference.

Note: I am definitely going to support Shiny Things's book for his contributions and efforts.

------------------------------------------------------------------------------------------------------------------------

The new Shiny Things Seal of Approval list:
* Singaporean stocks: ES3
* Singaporean bonds: A35
* Global stocks: IWDA on the London Stock Exchange (was VWRD, can also add in EMIM)
* Global(ish) bonds: QL2 (*Changed to CORP if you have >$100K)
* Broker for >$100k USD: Interactive Brokers
* Broker for <$100k USD: Standard Chartered

------------------------------------------------------------------------------------------------------------------------'

IWDA is optimized replication ETF and doesn't have EM. VWRD is fully replicated ETF and covers EM. These two fully justify 0.05 expense difference IMO

VWRD does indeed include about 1000 extra holdings and can probably expect a smaller tracking error though optimisation might lower the costs for the fund. About 400 of the extra holdings are EM stocks.

As for IWDA, the ETF holds 1621 different companies, while the MSCI World index which is tracked by IWDA constitutes of 1636 companies (as of december 31st 2014). It is not full replication, so if you are worried about those 15 companies having a major impact on your portfolio, do not bother with this ETF. I think the difference is too small to make a big deal out of it. It is physical replication (as mentioned in the fact sheet) instead of synthetic replication, which means they do hold the actual assets and not for instance derivatives as with synthetic replication. It is optimised in terms of not holding all the assets their benchmark holds, but as said earlier this paragraph, it comes close :wink: .

The FTSE index that VWRL/VWRD tracks is indeed more geographically diversified. If this is very important to you, you could add an emerging markets ETF or just buy the VWRL/VWRD ETF, it is your pick. I would go for IWDA+EMIM because I prefer that they are accumulating and having a really slight advantage in TER.

------------------------------------------------------------------------------------------------------------------------

Nah. If you care about emerging markets, then just go for VWRD; if you don't, then use IWDA. Using a whole bunch of little funds to fill in any gaps in IWDA is just going to cause you to run up transaction and rebalancing costs. Just pick one or the other.

Because they're issued by shitty issuers. Hyflux didn't even bother to get a credit rating on their bonds, they just stuffed the market full of these bonds on the assumption that "oh people have heard of us, we'll spin them a nice story, they won't care that we're heavily indebted and having trouble paying everything".

Bank bonds are good, especially Singaporean bank bonds, because those are rock-solid names. Bank pref shares are good if you're really gagging for income, though they should only be a small part of your portfolio. But high-yield long-dated junk-bond trash doesn't deserve a place in your portfolio, except through a well-diversified high-yield bond ETF - and even then it should be 3-5% of your portfolio, absolute max.

------------------------------------------------------------------------------------------------------------------------

STOCKS (70%)
1. 45% ES3
on SGX / SGD (SPDR® Straits Times Index ETF) / TER 0.30% / Semi-Annual Distribution

2. 20% IWDA
on LSE / USD (iShares Core MSCI World UCITS ETF) / TER 0.20% / No Distribution ie Reinvest dividends

3. 5% EIMI
on LSE / USD (iShares Core MSCI Emerging Markets IMI UCITS ETF) / TER 0.25% / No Distribution ie Reinvest dividends

BONDS (30%)
1. 20% A35
on SGX / SGD (ABF SINGAPORE BOND INDEX FUND) / TER 0.25% / Annual Distribution

------------------------------------------------------------------------------------------------------------------------

Hang on, hang on. The reason you glide down to bonds as you get older is so that you don't get clobbered by an equity downturn just before you retire. You make a tradeoff - giving up returns to reduce your risk. When you're younger, you can afford to take more risk and invest more of your money in stocks, because you can sit there and ride out a market cycle. But when you're older, you can't afford to get blindsided by a stock-market collapse, so you move into bonds, which have less volatility and smaller drawdowns.

You should have made most of your returns already by the time you're close to retiring; moving into bonds locks in those gains, as well.

On the one hand, you're saying "wow switching to bonds is a terrible idea", but then down below you're saying "what if you'd retired in 2009?". If you'd retired in 2009 with an all-equity portfolio, you'd have been completely ****ed, because your retirement portfolio had just taken a 50+% hit. But if you'd retired in 2009 with a 50-50 stocks-bonds portfolio, you'd have been pretty much fine. You might've taken a 20% drawdown, tops.

That right there is why you move into bonds when you get older. People who left their portfolios 100% in equities because "stocks have better returns, stocks will never go down!" were the ones who had to postpone their retirement because half their retirement portfolio got wiped out.

------------------------------------------------------------------------------------------------------------------------

If you keep waiting, you'll end up never buying. When it's going down, you'll think "oh no it's going down I don't want to buy it when it's going down!", and when it's going up you'll think "oh man it's going up, I'll wait for it to pull back a bit" and then you'll end up buying at the ding-dong high.

Don't be that guy.

Just put it in now and leave it for 20 years. The STI is relatively cheap at the moment (its price-to-earnings ratio is on the low side), so now's a good time to buy in if you're in for the long term.

------------------------------------------------------------------------------------------------------------------------

On agents:

It's good to remember that there's a conflict of interest there. The more expensive products they sell to you, the more money they make. If they have a product that gives you 3% returns and gives them 10% commission, vs a product that gives you 10% returns and 3% commission, be sure that they will sell you the first one and pocket their 10% commission.

They can still say, why not use your CPF? Extra 0.5% returns!

------------------------------------------------------------------------------------------------------------------------

On Algorithm Trading

------------------------------------------------------------------------------------------------------------------------

I have the minimum required to open an interactive brokers account and am planning to do 1000SGD of DCA in IWDA each month. Is the cost savings of spread premium + TT charges (each month) + account inactivity fee enough to warrant one to open an interactive brokers account since it seems that the TT charges (each month) seems to eat up the cost savings?

Yep, it's worth it. No wire charges, and FX at mid-market - the difference between Stanchart and IBKR for international stocks comes from the inactivity fee at IBKR, and SGD 1k a month is the line between "it's cheaper to pay Stanchart's FX rates" and "it's cheaper to pay IBKR's activity fee". And besides, you'll be at the $100k line in IBKR before long, which is where they stop charging the inactivity fee.

The new rule - if you've got more than $10k USD, and you do more than $500 USD a month, you should use IBKR. Above that amount, you'd be paying more in FX spread at Stanchart than you'd pay in maintenance fees at IB.

------------------------------------------------------------------------------------------------------------------------

To any kind of "would XXX stock rise", "will fed rise rates" kind of question:

I would be rich if I know.

------------------------------------------------------------------------------------------------------------------------

You do have FX exposure, but not quite where you think: you have FX exposure to the currency of the assets inside the ETF, not the currency of the ETF itself.

Let's imagine a hypothetical Zimbabwean ETF, listed in America. So the ETF is denominated in USD; it owns a bunch of Zimbabwean stocks valued in ZWD; and your home currency is SGD.

Let's say you buy $10,000 SGD worth of that ETF.

If the SGD halves in value, but everything else stays constant: your $10k SGD worth of the ETF is now worth $20k.

If the ZWD halves in value, but everything else stays constant: your $10k SGD worth of the ETF is only worth $5k (because the value of the assets in the ETF has halved).

If the USD halves in value: nothing happens! The value of the assets in the ETF has doubled in USD terms; but the value of the shares has halved in SGD terms.

------------------------------------------------------------------------------------------------------------------------

OK, here's the deal. Currency diversification in ETFs really doesn't matter - trust me on this, I can show you the math if you want - but if you try to buy a bunch of different currencies' ETFs while you're at Stanchart, you will get absolutely incinerated on FX costs. Stanchart's FX spreads for anything other than USD and SGD-denominated stocks are way too wide; this is my one and only complaint about them, but it's a big one.

You can move money directly in and out of IB. The really nifty thing if you're an expat is that you can move money in and out in multiple countries - so I can transfer money in to IBKR from my Singapore or Aussie bank accounts, convert it to USD (at interbank rates!), and then wire it out to my US bank account.

If you're doing a big lump of cash it can be a monster saving - imagine you've just sold a house in London for a million quid, and you need to convert the cash to USD. If you do it at the bank they'll probably take 30 pips out of you, that's $3,000 USD; but if you do it at Interactive you'll pay one or two pips, so you save nearly three grand.

At Stanchart, the spread on GBPSGD is about 2% wide - so you have to pay about 1% (half a spread) in foreign-exchange spread each time you want to buy or sell a GBP-denominated stock using SGD.

IBKR doesn't have a direct GBPSGD pair - you have to do "sell SGD for USD" and then "sell USD for GBP". But even though you have to do two trades instead of one, the spread only ends up being 0.02% - one-one-hundredth of what you'd pay at Stanchart.

------------------------------------------------------------------------------------------------------------------------

On the Permanent Portfolio:

I'm gonna set my stake in the ground right here: I think the Permanent Portfolio has some good ideas (diversify and rebalance), but I think the asset mix it chooses is sort of ridiculous, and it's not a good long-term investment plan because of that asset mix.

For anyone who hasn't run across it: the PP advocates 25% in long bonds, 25% in cash, 25% in gold, and 25% in stocks. Their argument is that it covers every possible economic and inflationary scenario: if the economy strengthens, bonds and stocks do well; if the economy weakens, cash and gold do well. If inflation, stocks and gold do well; if deflation happens, cash and bonds do well.

There are two huge problems with this, though.

The big one is that they're basically positioning for the apocalypse. Stocks do the best out of all asset classes over the very long term, so you want your portfolio mostly in stocks. These guys are one-quarter in stocks; I reckon if you looked at this portfolio since 1980 it would have been incinerated by a boring 60-40 stocks-bonds portfolio. Those allocations to cash and gold are absolutely huge, so when stocks and bonds outperform cash and gold (which is what happens the most often) you're going to have a huge drag on your portfolio from all the cash and gold you're sitting on.

Secondly, their 25-25-25-25 allocation doesn't reflect how the real world works. They've got 50% of their portfolio in stuff that works well during deflation, but (outside of Japan) deflation just doesn't happen that often. Same for economic shrinkage - they've got 50% of their portfolio in stuff that works when the economy's shrinking, but economies just don't shrink that often. If you wanted to weight this portfolio toward the actual probabilities of this stuff happening, you'd have maybe 5% of your portfolio in deflation hedges, and 10% in economic-shrinkage hedges... and that still leaves 85% to put in stocks and bonds, like a normal person.

The permanent portfolio is pretty much one step above allocating your portfolio to guns and canned goods.